SafeSpeed wrote:
Thanks for the answer. I was more thinking about the real-time risk management tasks. Have a look at this page:
http://www.safespeed.org.uk/sss.html - it gives a fair description of the sort of risk model that we all use subconsciously and continuously when we're driving.
Trying to control risk with a single parameter in isolation is doomed to fail. (Imagine for example deciding to paint the windscreen black and fitting a really big speedo.)
Oh, that's pretty straightforward. It's the old observe-evaluate-adjust thing. If no-one else was taught that, I highly recommend my old driving instructor. As Steve Haley states, it is "very simple" scientifically for easy digestion. I'm presuming the current form is for representation of the risk in a relatively text-book situation, for example trunk roads and fenced clearways.
Of course, there are many other factors in, for example, urban scenarios:
The speed is pretty much a constant, insomuch that the final decision feeds back into the equation, which revolves around this single factor (as far as it is the only factor the driver has control of).
The driver has some modicum of control over the space, but this does become an almost random variable in certain situations and can tend to zero in milliseconds (the child chasing the ball, the car not stopping at a blind drive/junction). Of course the distinction is whether we start calling the defintition 'Risk' or 'Blame', and even then we could also factor in consequence. Obviously the consequence of hitting a car that pulled out in front is very different at 20mph than at 40mph. I've seen it happen. It ain't pretty.
The Surprise element is of course the most erratic, as it can vary not only in nature, but also can on occasion be directly proportional to the possibly random nature of the space variable.
If we look at a combination of Risk and Consequence, then as the situation becomes more 'crowded' (Space and vision) then certain variables will have to be treated as the lowest common value that can reasonably be expected (someone will one day pull/run out in front of you with no warning) to minimise the sum of these.
I would say the only missing variable is Response, which of course is hugely variable too. you sneeze, this changes hugely for around 1-2 seconds.
Of course, all of this presumes an impeccably maintained vehicle with the driver fully alert, which I would hope apllies to us all.
I'm actually sure that this is similar to the type of calculation used to determine urban speed limits, must have a look around for it. In fact, I think in text book situations there is actually an inverse proportion of the factors, thus why the "only a fool breaks the two second rule" works.
I wouldn't mind looking into more of Steve's work as well, actually. I used to be a bit of a physics and mathematics geek when I was younger, and real-life applications interest me. Do you have further information on this site or does he have his own site?
Overall, we don't seem to disagree on much here. I'm quite surprised you're disagreeing with me. This equation in itself demonstrates that urban situations are vastly different from 'open road' / Outrun situations.
But, as I said, there are many more factors than speed anyway. Speed, while not always a causal factor, can be a major Risk factor (if taken in the context of 'Consequence Risks').
However, if you don't look at the road, you will crash. If you're watching the road, you wont stop for that car pulling out anyway. So then you have to think about consequence.
Phew!
